2003, Volume 5, Issue 3
Strategic Study of CAE >> 2003, Volume 5, Issue 3
A Multi-agent Model Based on Cellular Automata for Simulating Human Behavior to Assess Building Design
State Key Laboratory of Fire Science, University of Science and Technology of China, Hefei 230026, China
Next Previous
Abstract
In recent years, Cellular Automata (CA) micro-simulation has emerged as an effective technique for modeling traffic flow and achieved a lot of success. But the human movement and occupant evacuation are much more complex than traffic flow. It is necessary for us to consider more human behavior characteristics if try to apply the model in practice. For the advantages of agent in simulating human intelligence, a multi-agent model based on CA is presented which focuses on simulating the human behavior and the effect on human by building structure. In the model, people are represented by agents and they move over the network which is represented by a lattice of cells. Agents can perceive the local environment and chcx)se their moving direction autonomously. In order to simulate human movement in very complex structure building the acquainting and evacuating states based on the ideas of artificial life are introduced and that is proved successful in simulation. Using this model the effect of the building structure on human behavior can be studied and the performance of building design can be assessed.
Keywords
Cellular automata model ; multi-agent ; human behavior ; performance-based design
Figures
图1
图2
图3
References
[ 1 ] ] KaiN , MichaelS .Acellularautomatonmodelforfreewaytraffic[J].JPhysI, 1992, 2 (12) :2221～2229[2] FukuiM , IshibashiY .Trafficflowin1Dcellularautomatonmodelincludingcarsmovingwithhighspeed[J].JPhysSocJpn, 1996, 65 (6) :1868～1870
[ 3 ] RickertM , NagelK , SchreckenbergM , etal.Two lanetrafficsimulationsusingcellularautomata[J].PhysicaA , 1996, 231 (4) :534～550
[ 4 ] SimonPM , GutowitzHA .Cellularautomatonmodelforbi directionaltraffic[J].PhysicalReviewE , 1998, 57 (2) :2441～2444
[ 5 ] BihamO , MiddletonAA , LevineD .Self organizationanddynamicaltransitionintraffic flowmodels[J].PhysicalReviewA , 1992, 46 (10) :6124～6127
[ 6 ] ChowdhuryAD , SchadschneiderA .Self organizationoftrafficjamsincities:effectsofstochasticdynamicsandsignal periods[J].PhysicalReviewE :RapidCommunications (AIP , USA) , 1999, 59 (2) :1311～1314
[ 7 ] 谢惠民.复杂性与动力系统[M ].上海:上海科技教育出版社, 1994 link1
[ 8 ] BlueVJ , AdlerJL .Cellularautomatamicrosimulationformodelingbi directionalpedestrianwalkways[J].ForthcominginTransportationResearchB METH , 2001, 35 (3) :293～312
[ 9 ] YangLizhong, FangWeifengetal.Studyonoccupantevacuationinfirebasedoncellularautomatamodel[J].ChineseScienceBulletin, 2002, 47 (17) :1484～1488 link1
[10] MaesP .Agentsthatreduceworkandinformationoverload[J].CommunicationsoftheACM , 1994, 37 (7) :31～40
[11] WooldridgeMJ, JenningsNR .Intelligentagents:theoryand practice[J ].KnowledgeEngineeringReview.1995, 10 (2) :115～152
[12] SandipS .Multiagentsysterms:milestonesandnewhorizons[J].TrendsinCognitiveSciences.1997, 1 (9) :334～340
[13] BursteddeC , KlauckK , SchadschneiderAetal.Simulationofpedestriandynamicsusingatwo dimensionalcellularautomaton[J].PhysicaA , 2001, 295 (34) :507～525
[14] HelbingD , FarkasLandVicsekT .Simulatingdynamicalfeaturesofescapepanic[J].Nature, 2000, 407 (28) :487～490
京公网安备 11010502051620号
